Reading · by Apple

Apple Vision

Cut subjects out of photos and read body poses, offline.

image → text, data · image, document → data, text · image → data
Way in
Official APINo sign-up, no key
To start
Freefree, ships with macOS 27
Runs
On your Mac
Checked
Nov 3, 2026Prices as of that day

What it can do 7

Recognize text in images (OCR)first call ~26 s (model load), later calls fast; measuredimage → text, datafree (on-device, included with macOS)
Parse a document image into paragraphs, lists, tables and detected data (dates, money, phone)0.13 s/page after warm-up (measured)image, document → data, textfree (on-device, included with macOS)
Classify image contents and score aestheticsimage → datafree (on-device, included with macOS)
Detect faces, face landmarks, capture quality, humans, body/hand/animal poseimage, video → data, placefree (on-device, included with macOS)
Generate person, foreground-instance and saliency masks; promptable point/box/scribble segmentationimage → maskfree (on-device, included with macOS)
Read barcodes/QR, detect rectangles, documents, horizon, contours, lens smudge; track objects and optical flowimage, video → data, textfree (on-device, included with macOS)
Compute image feature prints for similarity searchimage → embeddingfree (on-device, included with macOS)

Does the same job

Prices and limits as of Nov 3, 2026, from Apple’s own pages. They change; check before you rely on them.Not affiliated with Apple. Not advice.Something wrong?
Made by Jasper Burns

A short note when I ship something new, about once a month. One list for all my projects.